Roy Lewallen wrote in news:13uqriie50k822
@corp.supernews.com:

....
I dug out a model of a 40/20 meter trapped dipole using traps using
RG-58 I did some time ago, and found the trap voltage to be 568 volts
RMS with 100 watts at 14.0 MHz. That's 800 volts peak. A model of a
40/20 meter trapped dipole with trap component X = 1000 ohms at
resonance and moderate inductor Q showed highest voltage of 648 RMS
(over 900 volts peak) at 7 MHz. So I'd want to use capacitors with a 2
kV rating to provide some margin.

I have done some mathematical modelling of so called coax traps with the
bootstrap connection (or the Hi Z connection as used in ARRL pubs), and
the work has halted needing some reconciliation with quality measurements
of inductor Q and trap impedance. The problem relates to estimating the
equivalent resistance in an inductor made from coax braid and covered in
the PVC jacket.

I see W8JI reports measurement of a coax trap at 7Mhz with about 18k ohms
at resonance, I have some data from DG1MFT who measured a trap around
7MHz on a R&S ZVRE VNA at somewhere around 22k ohms at resonance. I have
measured some prototypes at over 15k ohms using the TAPR VNA, but it is
not in the same class as quality instruments. In conclusion, I suspect
that it is likely that coax traps have impedance well over 10k ohms at
resonance.

One way to reduce the voltage impressed on the trap is to design the
system so that the trap is not close to resonance at any operating
frequency.

Such a design means moving beyond the simplistic explanation that such
antennas depend on traps acting like an on/off switch at the resonant
frequency, and that when they are resonant, the outboard conductors do
not exist.

Owen